[ODE] Re: Simulating Wheels

[Amund Børsand]

Steve Baker <sjbaker1 at airmail.net> skreiv:


> Well it *is* a problem if you are trying to reproduce the handling of
> a specific car (as I do) - and you want to predict the effect of doing
> things like putting lowering springs onto the car or putting more viscous
> oil in the shocks.

Well, long before that you have to get the center of mass correct for
the chassi, as well as correct mass and inertia for wheels, springs,
dampers, whishbones, spring struts, and positioning joints for the
suspension in correct places. If you get that measured up correctly you
WILL get a pretty realistic simulation... including all kinds of
suspension goodies like variable camber angles and stuff. Combine this
with Pacejka formulas, and a proper set-up steering rack and correct
camber, caster and toe-in, feed the feedback from the steering rack back
to a force-feedback steering wheel, and congratulations, you'll probably
have the most realistic car simulator ever! :) (This is what I'm
planning to do, don't steal my idea now :)

Oh, check the ODE manual. It has formulas for converting SI units of
spring rates and damping rates into CFM and ERP. That IS no problem ;) 
You can also find data on different springs and shocks on the net..
however it's not very easy to find. (Just like how do you know the
height of the center of gravity of your car? That's a tough one - which
HAS to be correct to make the car handle correctly.)

-- 
Amund Børsand <amund £ offroad . no>